Mechanistic studies of ethylene and α-olefin co-oligomerization catalyzed by chromium-PNP complexes

Article abstract of DOI:10.1021/om300492r

To explore the possibility of producing a narrow distribution of mid- to long-chain hydrocarbons from ethylene as a chemical feedstock, co-oligomerization of ethylene and linear α-olefins (LAOs) was investigated, using a previously reported chromium complex, [CrCl ₃(PNP^OMe)] (1, where PNP^OMe = N,N-bis(bis(o-methoxyphenyl)phosphino)methylamine). Activation of 1 by treatment with modified methylaluminoxane (MMAO) in the presence of ethylene and 1-hexene afforded mostly C₆ and C₁₀ alkene products. The identities of the C₁₀ isomers, assigned by detailed gas chromatographic and mass spectrometric analyses, strongly support a mechanism that involves five- and seven-membered metallacyclic intermediates comprised of ethylene and LAO units. Using 1-heptene as a mechanistic probe, it was established that 1-hexene formation from ethylene is competitive with formation of ethylene/LAO cotrimers and that cotrimers derived from one ethylene and two LAO molecules are also generated. Complex 1/MMAO is also capable of converting 1-hexene to C₁₂ dimers and C₁₈ trimers, albeit with poor efficiency. The mechanistic implications of these studies are discussed and compared to previous reports of olefin cotrimerization.

Full text of DOI:10.1021/om300492r

Article
pubs.acs.org/Organometallics
Mechanistic Studies of Ethylene and α-Olefin Co-Oligomerization
Catalyzed by Chromium−PNP Complexes
Loi H. Do, Jay A. Labinger,* and John E. Bercaw*
Arnold and Mabel Beckman Laboratories of Chemical Synthesis, California Institute of Technology, Pasadena, California 91125,
United States
S
* Supporting Information
ABSTRACT: To explore the possibility of producing a
narrow distribution of mid- to long-chain hydrocarbons from
ethylene as a chemical feedstock, co-oligomerization of
ethylene and linear α-olefins (LAOs) was investigated, using
a previously reported chromium complex, [CrCl3(PNPOMe)]
(1, where PNPOMe = N,N-bis(bis(o-methoxyphenyl)-
phosphino)methylamine). Activation of 1 by treatment with
modified methylaluminoxane (MMAO) in the presence of
ethylene and 1-hexene afforded mostly C6 and C10 alkene
products. The identities of the C10 isomers, assigned by
detailed gas chromatographic and mass spectrometric analyses, strongly support a mechanism that involves five- and seven-
membered metallacyclic intermediates comprised of ethylene and LAO units. Using 1-heptene as a mechanistic probe, it was
established that 1-hexene formation from ethylene is competitive with formation of ethylene/LAO cotrimers and that cotrimers
derived from one ethylene and two LAO molecules are also generated. Complex 1/MMAO is also capable of converting 1-
hexene to C12 dimers and C18 trimers, albeit with poor efficiency. The mechanistic implications of these studies are discussed and
compared to previous reports of olefin cotrimerization.
Scheme 1. Proposed Catalytic Cycle for the Selective
Trimerization of Ethylene to 1-Hexene
INTRODUCTION
■
Ethylene, the most widely produced organic compound in the
world,1,2 is attractive as a cheap and abundant feedstock for the
manufacturing of value-added products including linear α-olefins
(LAOs), a class of commodity chemicals that are used in
industries such as synthetic polymers, detergents, plasticizers,
and lubricants and have a global demand of over 4 million tons
per year.3 LAOs are produced primarily by catalytic oligomeriza-
tion of ethylene via linear chain growth (Cossee−Arlman
mechanism).4 Such reactions produce a nonselective Schulz−
Flory distribution of oligomers, requiring additional procedures
to separate the desired products. More recently, selective
oligomerization of ethylene to 1-hexene and/or 1-octene has
been reported,5−8 first using chromium/2-ethylhexanoate9,10
and subsequently with complexes containing a diverse array of
transition-metal ions and ligand architectures.5 These reactions
most likely proceed through a metallacycle ring expansion
mechanism, in which oxidative coupling of two ethylene
molecules at an electrophilic metal center affords a metalla-
cyclopentane intermediate, which undergoes ethylene insertion
to give a metallacycloheptane that eliminates 1-hexene (Scheme
1).11−16 The elimination step may involve either a β-hydride
elimination/reduction elimination sequence or a concerted 3,7-
hydride shift; it is difficult to distinguish between these
alternatives,13 but the latter is favored by recent theoretical
studies.15−17 Whichever mechanism operates, the elimination is
expected on geometric grounds to be much more favorable,
relative to further insertion, for the seven-membered rather than
for the five-membered metallacycle, accounting for the observed
selectivity. With some catalysts, and especially at higher ethylene
pressures, a second insertion becomes favored over elimination,
yielding 1-octene. This chemistry has also been extended to
homotrimerization of LAOs18,19 and other olefinic sub-
strates.20−22
Significant effort has been devoted to understanding the
chemistry of selective ethylene oligomerization,5−7 including
Received: June 1, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/om300492r | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
issues such as catalyst initiation, effect of activators, metal
oxidation states, and reaction mechanism. One interesting
feature of ethylene trimerization is the formation of some C10
alkenes in addition to 1-hexene, suggested to be secondary
products from the coupling of two ethylenes with 1-hexene. Co-
oligomerization of ethylene and an α-olefin, such as propylene,13
1-octene,23 and styrene,20 by homogeneous chromium catalysts
has been demonstrated previously. “Product recycling”, wherein
co-oligomerization of ethylene with the primary α-olefin
produced competes effectively with homo-oligomerization,
could constitute a selective route to hydrocarbons larger than
C6 or C8, using only ethylene as starting material.24 Here we
explore the feasibility of such a process by examining the
efficiency of 1-hexene incorporation in cotrimerization, using the
previously reported chromium complex [CrCl3(PNPOMe)] (1,
where PNPOMe = N,N-bis(bis(o-methoxyphenyl)phosphino)-
methylamine; Chart 1) as the catalyst precursor.25,26
and C18 (0.2 wt %) olefins (Figure S1B). Because 1-hexene was
used as a starting material, the amount that was formed during
the reaction could not be determined; the productivity based on
the other products was 926 93 g/((g of Cr) h) (Table 1, entry
2).
The C10 olefins were isolated by fractional distillation of the
crude reaction mixture. GC analysis showed three major species
with retention times of 36.98, 37.19, and 37.38 min and several
minor ones between 37.75 and 38.07 min (Figure 1, black trace).
By comparison to authentic samples of the nine isomers expected
from cotrimerization through a metallacycle mechanism
(Scheme 2), it is apparent that 3-propyl-1-heptene (P, red
trace) corresponds to the peak at 36.98 min, 4-ethyl-1-octene (O,
orange trace) corresponds to the peak at 37.19 min, 5-methyl-1-
nonene (N, yellow trace) corresponds to the peak at 37.38 min,
2-butyl-1-hexene (M, green trace) corresponds to the peak at
37.75 min, and the linear C10 olefins (H−L; light blue, brown,
pink, purple, and blue traces) correspond to the peaks between
37.78 and 38.07 min. The relative amounts of C10 olefins were
determined from the integrated areas of the peaks in the gas
chromatogram: linear C10 olefins H−L, 3%; M, 5%; N, 47%; O,
18%; P, 27% (Scheme 2).
Chart 1. Structure of the Chromium Precatalyst
[CrCl3(PNPOMe)] (1)
To obtain further support for the assignments, a sample of the
C10 olefin mixture was hydrogenated and analyzed by GC. As
shown in the black trace in Figure 2, the gas chromatogram of the
saturated C10 products displays two prominent peaks at 36.79
and 36.90 min and a significantly smaller peak at 37.72 min. A
comparison of the retention times of the peaks in this trace to
those of the expected products (Scheme 2) indicates that, indeed,
they correspond to 2-ethyloctane (S, red trace), 5-methylnonane
(R, orange trace), and decane (Q, green trace), respectively, with
a relative ratio of 40:56:4 (Figure 3). (Although the expected
ratio based on the distribution of C10 alkenes is 45:52:3, this
discrepancy is attributable to experimental error, such as
nonideal peak integration, slight differences in the GC response
factors for the different isomers, or incomplete hydrogenation.)
Consistent with the chemical formula C10H22, mass spectral
analysis reveals that all three species in the hydrogenated sample
display a peak with an m/z value of 142. In addition, each
compound exhibits a unique fragmentation pattern in its mass
spectrum that matches well with that of the species to which it
was assigned (Figure 3).
RESULTS
■
Co-Oligomerization of Ethylene and 1-Hexene. To
establish a benchmark for the present studies, compound 1 was
treated with modified methylaluminoxane (MMAO) in chloro-
benzene under an atmosphere of ethylene. Analysis of the
reaction mixture after 1 h by gas chromatography (GC) (Figure
S1A, Supporting Information) revealed that C6 (92 wt %) and
C10 (8 wt %) alkenes were the primary products (Table 1, entry
1), with a productivity of 571 14 g/((g of Cr) h); 1-hexene was
the only detectable C6 isomer, and only trace amounts (<10 mg)
of polyethylene were obtained.
Homo-Oligomerization of 1-Hexene. To test the
feasibility of LAO homo-oligomerization by the Cr-PNPOMe
system, reactions were conducted using 1/MMAO in the
presence of 1-hexene without added ethylene. GC-MS analysis
of this reaction mixture showed that both C12 and C18 olefins
were formed, in a weight percent ratio of 27:73 (Figure S2,
Supporting Information; Table 1, entry 4). Although several
isomers of each chain length were obtained, no attempt was
To favor the production of C10 olefins and to obtain sufficient
material for characterization, the same reaction was performed
with an excess of added 1-hexene; the ratio of 1-hexene to
ethylene in solution under the reaction conditions employed, on
the basis of NMR spectroscopic measurements, was approx-
imately 1:1. Quantification of olefins heavier than C6 by GC
showed that C10 olefins were the major products (94 wt %),
followed by much smaller amounts of C12 (1 wt %), C14 (4 wt %),
a
Table 1. Product Distribution of Olefin Homo- and Cotrimerization
product (wt %)
e
f
entry
monomer(s)
productivity (g/((g of Cr) h))
C6
C10
C11
43
C12
C13
C14
C15
0.8
C16
1.2
C18
b
1
2
3
4
ethylene
571 14
926 93
863 97
7.6 0.3
92
nd
51
8
94
2
b
c
ethylene /1-hexene
1
4
0.2
73
b
d
ethylene /1-heptene
0.5
0.6
c
1-hexene
27
a
Reaction conditions: complex 1 (12 μmol), monomer(s), and MMAO (250 mg, ∼55 equiv, solution in isohexanes) diluted to 10 mL with
b
c
d
e
chlorobenzene, room temperature. Ethylene (1 atm). 1-Hexene (35 mmol). 1-Heptene (24 mmol). The standard deviations were determined
from reactions performed in triplicate. The C6 fraction contains >99% 1-hexene.
f
B
dx.doi.org/10.1021/om300492r | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 1. Gas chromatogram of the C10 alkenes (black, bottom trace) obtained by fractional distillation from the reaction of 1/MMAO/ethylene/1-
hexene. The products appear at 36.98, 37.19, 37.38, 37.75, 37.86, and 38.07 min. The C10 olefin standards are shown directly above (from top to bottom,
RT = retention time in min): cis-5-decene (H, light blue; 38.08); cis-4-decene (I, brown; 38.07); 1-decene (J, pink; 37.84); trans-4-decene (K, purple;
37.86); trans-5-decene (L, blue; 37.90); 2-butyl-1-hexene (M, green; 37.76); 5-methyl-1-nonene (N, yellow; 37.44); 4-ethyl-1-octene (O, orange;
37.22); 3-propyl-1-heptene (P, red; 37.02). The peak marked with an asterisk (*) is the trans-4-decene impurity present in the 3-propyl-1-heptene
standard (see the Supporting Information).
a
Scheme 2. C6 and C10 Alkenes Formed via Cotrimerization of Ethylene and 1-Hexene and Their Hydrogenation Products
a
The percentages shown in green correspond to the relative amounts of C10 alkenes present in the reaction product. Similarly, the percentages
shown in blue correpond to the relative amounts of C10 alkanes present after the alkenes were subjected to hydrogenation. The discrepancies (up to
5%) between the total amount of each group of alkene isomers observed and the amounts of hydrogenation products obtained are attributed to
experimental error.
made to ascertain their identities. The overall productivity was
quite low: 7.6 0.3 g/((g of Cr) h).
Co-Oligomerization of Ethylene and 1-Heptene. An
analogous reaction was carried out with ethylene and 1-heptene
C
dx.doi.org/10.1021/om300492r | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
occur.28−30 Although the methoxyaryl functionalities are not
essential to obtaining a high ratio of 1-hexene versus 1-octene,31
only the PNPOMe ligand forms a catalyst that produces 1-hexene
with >99% trimerization selectivity.
The ability of 1 to facilitate selective trimerization is indicated
by the high yield of C10 olefins: ∼94 wt % of the quantifiable
products (i.e., those other than C6) obtained when ethylene/1-
hexene were used as comonomers (Table 1, entry 2). Because
identification of the reaction products would provide insight into
their possible mechanism of formation, studies were undertaken
to characterize the C10 species. As shown in Scheme 2, there are
three possible metallacycloheptanes from cotrimerization of two
ethylenes with one 1-hexene and two possible routes to each of
them. Unsubstituted chromacyclopentane A (from oxidative
coupling of two ethylene molecules with the Cr center) can
undergo insertion of ethylene to give D (i.e., homotrimerization)
or either 2,1- or 1,2-insertion of 1-hexene, leading to
chromacycloheptanes E and F, respectively. Alternatively,
chromacyclopentanes B and C, formed by oxidative coupling
of ethylene and 1-hexene, could undergo subsequent insertion of
another ethylene to yield chromacycloheptanes E and G or F and
G, respectively.
Elimination of olefin from the various chromacycloheptanes,
by either the stepwise or concerted route (see above), would
afford the products shown in Scheme 2. Only linear decenes can
result from E, with the position of the double bond depending on
the relative preference for hydride elimination (or transfer) from
substituted vs unsubstituted and endo- vs exocyclic Cβ positions.
Likewise, F should give rise to either 2-butyl-1-hexene (M) or 5-
methyl-1-nonene (N) and G to 4-ethyl-1-octene (O) or 3-
propyl-1-heptene (P). As shown in Figure 1, GC analysis shows
that all of the branched and some or all of the unbranched
expected C10 olefins are indeed formed, supporting the proposed
mechanism.
Figure 2. Gas chromatogram of the C10 alkanes (black, bottom trace)
obtained from hydrogenation of the C10 alkene mixture (see Figure 1).
The products appear at 36.79, 36.90, and 37.72 min. The C10 alkane
standards are shown directly above (from top to bottom, RT = retention
time in min): decane (Q, green; 37.28); 5-methylnonane (R, orange;
36.86); 2-ethyloctane (S, red; 36.71).
as comonomers, in an approximately 4:3 ratio (as determined by
NMR spectroscopic measurements). GC analysis of the crude
mixture showed that 1-hexene and C11 olefins constituted
approximately 51 and 43 wt %, respectively, of the total products
(Figure S1C, Supporting Information; Table 1, entry 3). Minor
amounts of C10 (2 wt %), C13 (0.5 wt %), C14 (0.6 wt %), C15 (0.8
wt %), and C16 (1.2 wt %) olefins were also observed. The overall
productivity was 863 97 g/((g of Cr) h).
DISCUSSION
■
The chromium(III) complex CrCl3(PNPOMe) (1) was selected
as the precatalyst for the present studies because of its high
selectivity for ethylene trimerization.25−27 That selectivity has
been attributed, in part, to the interaction of a pendant methoxy
moiety of the PNPOMe ligand with the chromium center;7 when
these methoxy groups are absent, as in the [CrCl3(PNP)]2
variant (where PNP = N,N-bis(diphenylphosphino)-
methylamine), both ethylene tri- and tetramerization
Detailed analysis of the C10 olefin distribution indicates several
important characteristics of the [Cr-PNPOMe] intermediates.
First, since only 3−4% of the C10 olefins are linear, their
precursor intermediate E must be significantly disfavored. E can
Figure 3. Electron-impact (EI) ionization mass spectra (black) of the peaks that appear in the C10 alkane sample (Figure 2) at (A) 36.79 min, (B) 36.90
min, and (C) 37.72 min. For comparison, the mass spectra for authentic samples of 4-ethyloctane (pink), 5-methylnonane (blue), and decane (green)
are shown directly above those for A, B, and C, respectively. RT = retention time.
D
dx.doi.org/10.1021/om300492r | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
be generated either by 2,1-insertion of 1-hexene into
chromacyclopentane A or ethylene insertion into the unsub-
stituted side of chromacyclopentane B; both sequences must
therefore be of low probability. In each case there are several
possible interpretations. A is certainly formed, since homotrime-
rization of ethylene is taking place (see below); hence, insertion
of olefin into A exhibits a strong preference for ethylene over
LAO and/or for 1,2- over 2,1-insertion. In previous studies using
a Cr-PNPOMe/biphenyldiyl complex as a model for the
chromacyclopentane intermediate,13 it was found that ethylene
inserts into the five-membered metallacycle more than 20 times
faster than LAOs and the latter insert exclusively in 1,2-mode,
consistent with the current observations. In contrast, in the
cotrimerization of ethylene and styrene by 1/MAO the major
products (∼80%) were those derived from 2,1-insertion of
unsubstituted one. Again, this pattern does not appear to be
general for all ethylene trimerization catalysts. For example,
ethylene trimerization by a titanium−phenoxyimine complex,
believed to involve a similar metallacycle mechanism, afforded M
as the major (∼90%) C10 olefin product.40 The more similar
yields of O and P (18% and 27%, respectively) imply that
proximity to a substituent in the γ position has a much smaller
effect on hydride transfer preference.
Although the formation of C10 olefins from the cotrimerization
of one 1-hexene and two ethylenes is unambiguous, routes to the
observed C12, C14, and C18 products (Table 1, entry 2) are less
obvious. The various possible combinations are shown in Table
2; the C14 and C18 olefins are most simply explained as secondary
Table 2. Possible Combinations of Ethylene (C2), 1-Hexene
(C6), and 1-Heptene (C7) Dimers and Trimers That Make Up
styrene into A, and no 1,2-insertion products were observed.20
A
a
preference for 2,1-insertion of styrene is common, both in the
polymer literature32−34 and elsewhere,35−38 although 1,2-
insertion can be achieved using sterically hindered catalysts.39
It seems quite unlikely that insertion of ethylene into the
substituted position of B, to give G, should be favored over
insertion into the unsubstituted position, to give E. A much more
reasonable explanation is that formation of B itself is relatively
unfavorable, presumably for steric reasons, and that isomer C is
strongly favored in the co-oxidative coupling of 1-hexene with
ethylene.
the Observed Product Distributions
b
ethylene
ethylene/1-hexene
ethylene/1-heptene
C2 + C2 + C2 = C6
C2 + C2 + C6 = C10
C2 + C2 + C2 = C6
C2 + C2 + C6 = C10
C6 + C6 = C12
C2 + C2 + C2 = C6
C2 + C2 + C6 = C10
C2 + C2 + C7 = C11
C6 + C7 = C13
C2 + C2 + C10 = C14
C2 + C6 + C6 = C14
C2 + C2 + C14 = C18
C6 + C6 + C6 = C18
C7 + C7 = C14
C2 + C2 + C10 = C14
C2 + C6 + C6 = C14
C2 + C6 + C7 = C15
C2 + C7 + C7 = C16
The relative overall rates of homo- and cotrimerization may be
inferred to be approximately the same, from the observation that
the reaction of ethylene with 1-heptene (at an approximately 4:3
ratio in solution) gives close to 50% each of 1-hexene and heavier
products (Table 1). Since there is more than one available route
to each intermediate, it is not possible to assess the relative
contributions of all the pathways. However, as the above model
studies show, insertion of LAO is significantly less favored than
that of ethylene. Hence, a reasonable interpretation would be
that A and C are formed at about the same rate, with little or no B,
and that both A and C react further almost entirely by insertion of
ethylene rather than LAO. A smaller amount (∼4%) of C14 and
even less (∼0.2%) C18 is formed in the reaction of ethylene with
1-hexene; these could result from sequential secondary reactions
of C10 and C14 with two ethylenes. Another possible route to C14
would be insertion of C6 into C; that is, a C2/LAO/LAO
cotrimerization. Results from the ethylene/1-heptene co-
oligomerization experiment suggest that the latter route does
contribute (see below). In a related study on co-oligomerization
of ethylene with 1-octene using a bis(pyridine)carbene complex
of Cr, insertion of LAO into the analogue of intermediate C was
indeed found to contribute to the products,23 although that
system exhibits several other significant differences from the
present one: there the C12 products (the result of cotrimerization
of two ethylenes with one 1-octene) did not predominate in the
product distribution, and the most abundant C12 species
identified was 2-butyl-1-octene, the analogue of M, which is
not the major C10 product here.
a
Each combination must satisfy the following criteria: (1) homo-
oligomerization of ethylene form trimers, (2) co-oligomerization of
ethylene/LAO forms trimers, (3) formation of dimers and trimers
from homo-oligomerization of LAO, (4) no formation of dimers
between LAO and ethylene, and (5) no formation of tetramers of
b
alkenes. Use of 1-heptene allowed the determination of trimers
formed from two molecules of LAOs and one molecule of ethylene.
and tertiary cotrimerization products of two ethylene molecules
with the C10 and C14 olefins generated during the reaction.
Cotrimerization of one ethylene with two 1-hexenes is an
alternate route to C14, as noted above and discussed further
below.
Because codimerization of ethylene/LAO does not occur, as
evidenced by the absence of any C8 products from ethylene and
1-hexene, C12 cannot be the coupling product of ethylene and
C10. A more reasonable route is through homodimerization of 1-
hexene; 1/MMAO was found to homo-oligomerize 1-hexene, at
a productivity about 1% of that of reactions involving ethylene
(Table 1, entry 4; Figure S2, Supporting Information). The yield
of C12 in the ethylene/1-hexene cotrimerization, about 1% of all
products heavier than C6, is roughly consistent with that relative
reactivity; some of the C18 products might also arise by that route.
In comparison, chromium complexes supported by triazacyclo-
hexane ligands (Cr−NNN) can oligomerize LAO with high
efficiency, up to ∼98% conversion;18,19 the bis(pyridine)carbene
Cr complex is also effective for homodimerization of LAOs.23
Some additional inferences may be obtained from the reaction
of ethylene/1-heptene with 1/MMAO, which affords C6, C10,
C11, C13, C14, C15, and C16 species (Table 1, entry 3). Clearly C6
and C11 arise via homotrimerization of ethylene and cotrimeriza-
tion of two ethylenes and one 1-heptene, respectively; because
they are formed in comparable amounts (51 and 43 wt %), the
reaction rates for homotrimerization and cotrimerization by 1
must be similar. (Although no attempts were made to measure
Insertion of ethylene into C can take place on either the side
near to or away from the β substituent to give G and F,
respectively. Since the sums of the products M + N and O + P are
not too different, it appears that there is relatively little preference
in that step. On the other hand, it is notable that N (47%) is
obtained in a significantly higher yield than M (5%), suggesting
that hydrogen transfer, whether it occurs in a stepwise or
concerted manner,15−17 is less favored from the tertiary
substituted β position in F than from the secondary,
E
dx.doi.org/10.1021/om300492r | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
actual reaction kinetics in the present work, it has been the
subject of other reports in the literature.41,42) The remaining
identified products comprise only a small percentage of the total.
The only obvious pathway to C13 is by codimerization of C6 and
C7; a small amount of LAO dimerization was indicated by
appearance of C12 in the ethylene/1-hexene reaction, but since
much more C7 than C6 is present at any time here, it is surprising
that the amounts of C14 and C13 are nearly the sameespecially
given that there are additional possible routes to C14 isomers,
cotrimerization of C2/C2/C10 (not likely to be very significant, as
the amount of C10 is relatively low) or C2/C6/C6. In contrast to
the situation for ethylene/1-hexene, where the two sequences
C2/C2/C10 and C2/C6/C6 both give C14 and hence their relative
contributions cannot be determined, the corresponding
sequences here would be C2/C2/C11 and C2/C7/C7, giving C15
and C16 respectively. Again, comparable amounts of the two are
obtained, implying that some cotrimerization of the form C2/
LAO/LAO does take place.
EXPERIMENTAL SECTION
■
General Considerations. Reagents obtained from commercial
suppliers were used as received. The chromium precatalyst
[CrCl3(PNPOMe)] (1) was synthesized according to a literature
procedure.12 The following gas chromatography standards were
obtained from the sources shown in parentheses: 4-ethyl-1-octene
(Novel Chemical Solutions), 2-butyl-1-hexene (ChemSampCo), trans-
5-decene (Aldrich), cis-5-decene, (TCI), trans-4-decene (ChemSamp-
Co), 1-decene (Aldrich), and decane (Aldrich). The compounds 5-
methylnonane and 4-ethyloctane were prepared by hydrogenation of 2-
butyl-1-hexene and 4-ethyl-1-octene, respectively, using H2 over Pd/C.
Syntheses of 3-propyl-1-heptene, 5-methyl-1-nonene, and cis-4-decene
are provided in the Supporting Information. The modified methyl-
aluminoxane activator (MMAO-C4 solution in isohexanes, 7 wt % Al;
referred to as MMAO in the text) was obtained from Albemarle.
MMAO-C4 was the activator of choice because it is available as a
homogeneous solution that could be transferred quantitatively in
repeated trials and gave the most reproducible data. Chlorobenzene was
purged with argon and dried over calcium hydride before use. All
ethylene oligomerization trials were performed using a high-vacuum
Schlenk manifold. Polymer-grade ethylene gas (>99.9% purity) was
purified by passage through columns containing activated molecular
sieves and Ridox, an oxygen scavenger.
Gas Chromatographic Analyses. Gas chromatography−mass
spectrometry (GC-MS) was performed using an Agilent 6890N system
with an HP-5MS capillary column (30 m length, 0.25 mm diameter, and
0.5 μm film) that was equipped with an Agilent 5973N mass selective
detector. Gas chromatography (GC) was performed using an Agilent
6890N instrument with a flame ionization detector (FID). Routine runs
were performed using a DB-1 capillary column (10 m length, 0.10 mm
diameter, 0.40 μm film) with the following heating program: hold at 40
°C for 3 min, ramp temperature at 50 °C/min to 290 °C and then hold
for 3 min (total run time 13 min). For detailed C10 isomer analyses, an
Agilent GS-GasPro capillary column (30 m, 0.32 mm diameter) was
used, with the following heating program: hold at 90 °C for 30 min,
ramp temperature at 50 °C/min to 260 °C, and then hold for 10 min
(total run time 43 min).
The amount of products in each oligomerization trial was determined
from the integrated areas of the peaks observed in the gas
chromatogram, using the integrated area of a biphenyl internal standard
as a reference. The integrated area for each peak was corrected using the
appropriate response factor, which was determined experimentally. The
response factors for all isomers containing the same number of carbon
atoms were assumed to be the same.
To determine the identities of the C10 alkene isomers, the C10 fraction
was separated from the crude reaction mixture by fractional distillation
under atmospheric pressure. The GC retention times of the C10
products were compared to those of authentic samples that were either
prepared independently or obtained from commercial sources. These
assignments were further confirmed by GC analysis of the C10 sample
after it had been hydrogenated.
Procedure for Olefin Oligomerization Trials. Homotrimeriza-
tion of Ethylene. Inside a nitrogen-filled glovebox, [CrCl3(PNPOMe)]
(8.0 mg, 12 μmol) was suspended in 10 mL of chlorobenzene, along
with a stir bar, in a 250 mL round-bottom flask equipped with a 180°
needle valve. The flask was sealed, brought outside of the glovebox, and
then attached to a high-vacuum manifold. The reaction flask was cooled
to −78 °C in a dry ice/acetone bath, degassed, and then warmed to
room temperature. The solution was stirred under an atmosphere of
ethylene, and a solution of MMAO (250 mg diluted to 1 mL in
chlorobenzene) was added using an airtight syringe through a septum
that was placed over the needle valve adapter, resulting in the formation
of a pale yellow-green homogeneous mixture. Ethylene consumption
was monitored using a U-tube mercury monometer during the course of
the reaction, and the ethylene pressure was maintained between 700 and
790 Torr. (It should be noted that gradual catalyst decomposition was
observed during the course of the reaction; the underlying cause for such
behavior has not yet been determined.) After 1 h, the reaction was
quenched with HCl(aq) and stirred for 1 min, resulting in a colorless
CONCLUSIONS
■
The following generalizations appear to best account for the
observations in co-oligomerization of ethylene and LAO by the
chromium−PNPOMe complex: (1) the initial intermediates are
chromacyclopentanes formed by oxidative coupling of either two
ethylenes or one ethylene and one LAO at the Cr center, (2) the
rate constants for formation of those two species are similar, (3)
the isomer of the mixed chromacyclopentane with the alkyl chain
in the β position is strongly favored, (4) insertion of ethylene into
the five-membered ring is strongly favored over insertion of
LAO, (5) insertion occurs preferentially on the side of the ring
away from the substituent, (6) hydride transfer occurs
preferentially from a secondary over a tertiary carbon site, (7)
homodimerization (and homotrimerization) of LAOs does take
place, but at a rate much lower than that of any reactions
involving ethylene, and (8) codimerization of ethylene with LAO
is not observed. While points 7 and 8 seem at first to be at odds,
they need not be: it may be that the disubstituted
chromacyclopentane is sufficiently inhibited toward a further
insertion that formation of dimer (presumably by slow β-
hydrogen elimination/reductive elimination) competes effec-
tively with further growth, whereas for un- or monosubstituted
rings (intermediates A−C in Scheme 2) ethylene insertion
dominates over elimination. Only small amounts of the products
that would result from insertion of ethylene into a disubstituted
ring (C16 from ethylene/1-heptene; probably some of the C14
from ethylene/1-hexene) are observed.
Comparison of 1 to other olefin trimerization catalysts in the
literature (see the Discussion) shows clearly that these
generalizations do not universally apply to reactions proceeding
via the metallacycle-based mechanism. There is a wide range of
reactivity behaviors and consequent product distribution
profiles, including preferences for ethylene vs LAO in both
oxidative coupling and insertion steps, relative rates of growth by
insertion vs elimination by hydride transfer, and regiochemistry
of insertion and hydride transfer. Ongoing work is aimed at
gaining the understanding of the factors influencing those
preferences, at a quantitative as well as qualitative level, that will
be needed to make this chemistry a viable alternative to
traditional nonselective oligomerization as a route from ethylene
to C10+ hydrocarbons.
F
dx.doi.org/10.1021/om300492r | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
solution. Solid biphenyl was added (50 mg, 325 μmol) as an internal
standard. A 2.0 mL aliquot of the organic layer was filtered through a
short silica plug and then analyzed by GC.
(12) Agapie, T.; Schofer, S. J.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem.
Soc. 2004, 126, 1304−1305.
(13) Agapie, T.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2007,
129, 14281−14295.
Co-Oligomerization of Ethylene/1-Hexene. The same procedure
was used as described for homotrimerization of ethylene, except that 1-
hexene (2.98 g, 35.4 mmol) was added to the reaction flask in addition to
the chromium precatalyst before diluting to 10 mL with chlorobenzene.
Co-Oligomerization of Ethylene/1-Heptene. The same procedure
was used as described for homotrimerization of ethylene, except that 1-
heptene (2.32 g, 24 mmol) was added to the reaction flask in addition to
the chromium precatalyst before diluting to 10 mL with chlorobenzene.
Less 1-heptene was added to the reaction mixture, in comparison to the
amount of 1-hexene used in the ethylene/1-hexene trials, because its
increased lipophilicity was found to cause clumping of the chromium
complex at high concentration when activated with MMAO-C4.
Homo-Oligomerization of 1-Hexene. Inside a nitrogen-filled
glovebox, [CrCl3(PNPOMe)] (8.0 mg, 12 μmol) and 1-hexene (2.98 g,
35 mmol) were diluted to 10 mL with chlorobenzene. The mixture was
treated with MMAO-C4 (250 mg) and then stirred at room temperature
for 1 h. Aqueous HCl was added to quench the reaction, and then
biphenyl (50 mg, 325 μmol) was added to the organic layer as an internal
GC standard. A 2.0 mL aliquot of the organic phase was filtered through
a short silica plug and then analyzed by GC.
(14) Emrich, R.; Heinemann, O.; Jolly, P. W.; Kruger, C.; Verhovnik,
̈
G. P. J. Organometallics 1997, 16, 1511−1513.
́
(15) Janse van Rensburg, W.; Grove, C.; Steynberg, J. P.; Stark, K. B.;
Huyser, J. J.; Steynberg, P. J. Organometallics 2004, 23, 1207−1222.
(16) Qi, Y.; Dong, Q.; Zhong, L.; Liu, Z.; Qiu, P.; Cheng, R.; He, X.;
Vanderbilt, J.; Liu, B. Organometallics 2010, 29, 1588−1602.
(17) Klemps, C.; Payet, E.; Magna, L.; Saussine, L.; Le Goff, X. F.; Le
Floch, P. Chem. Eur. J 2009, 15, 8259−8268.
(18) Kohn, R. D.; Haufe, M.; Kociok-Kohn, G.; Grimm, S.;
̈
̈
Wasserscheid, P.; Keim, W. Angew. Chem., Int. Ed. Engl. 2000, 39,
4337−4339.
(19) Wasserscheid, P.; Grimm, S.; Kohn, R. D.; Haufe, M. Adv. Synth.
̈
Catal. 2001, 343, 814−818.
(20) Bowen, L. E.; Wass, D. F. Organometallics 2006, 25, 555−557.
(21) Bowen, L. E.; Charernsuk, M.; Wass, D. F. Chem. Commun. 2007,
2835−2837.
(22) Bowen, L. E.; Charernsuk, M.; Hey, T. W.; McMullin, C. L.;
Orpen, A. G.; Wass, D. F. Dalton Trans. 2010, 39, 560−567.
(23) McGuinness, D. S. Organometallics 2009, 28, 244−248.
(24) Tomov, A. K.; Chirinos, J. J.; Long, R. J.; Gibson, V. C.; Elsegood,
M. R. J. J. Am. Chem. Soc. 2006, 128, 7704−7705.
ASSOCIATED CONTENT
■
(25) Carter, A.; Cohen, S. A.; Cooley, N. A.; Murphy, A.; Scutt, J.;
Wass, D. F. Chem. Commun. 2002, 858−859.
S
* Supporting Information
Text and figures giving syntheses of gas chromatography
standards and GC traces. This material is available free of charge
via the Internet at http://pubs.acs.org.
(26) Agapie, T.; Day, M. W.; Henling, L. M.; Labinger, J. A.; Bercaw, J.
E. Organometallics 2006, 25, 2733−2742.
(27) Schofer, S. J.; Day, M. W.; Henling, L. M.; Labinger, J. A.; Bercaw,
J. E. Organometallics 2006, 25, 2743−2749.
(28) Bollmann, A.; Blann, K.; Dixon, J. T.; Hess, F. M.; Killian, E.;
Maumela, H.; McGuinness, D. S.; Morgan, D. H.; Neveling, A.; Otto, S.;
Overett, M.; Slawin, A. M. Z.; Wasserscheid, P.; Kuhlmann, S. J. Am.
Chem. Soc. 2004, 126, 14712−14713.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: jal@its.caltech.edu (J.A.L.); bercaw@caltech.edu
(J.E.B.).
(29) Elowe, P. R.; McCann, C.; Pringle, P. G.; Spitzmesser, S. K.;
Bercaw, J. E. Organometallics 2006, 25, 5255−5260.
Notes
(30) Blann, K.; Bollmann, A.; de Bod, H.; Dixon, J. T.; Killian, E.;
Nongodlwana, P.; Maumela, M. C.; Maumela, H.; McConnell, A. E.;
The authors declare no competing financial interest.
́
Morgan, D. H.; Overett, M. J.; Pretorius, M.; Kuhlmann, S.;
ACKNOWLEDGMENTS
Wasserscheid, P. J. Catal. 2007, 249, 244−249.
(31) Blann, K.; Bollmann, A.; Dixon, J. T.; Hess, F. M.; Killian, E.;
Maumela, H.; Morgan, D. H.; Neveling, A.; Otto, S.; Overett, M. J.
Chem. Commun. 2005, 620−621.
■
This work was supported by BP. L.H.D. acknowledges the
National Institute of General Medical Sciences for a postdoctoral
fellowship (F32 GM099189-02). We thank Dr. Emmanuelle
Despagnet-Ayoub, Matt Winston, and Rachel Klet for helpful
comments on this paper.
(32) Pellecchia, C.; Pappalardo, D.; D’Arc, M.; Zambelli, A.
Macromolecules 1996, 29, 1158−1162.
(33) Correa, A.; Galdi, N.; Izzo, L.; Cavallo, L.; Oliva, L.
Organometallics 2008, 27, 1028−1029.
(34) Luo, Y.; Luo, Y.; Qu, J.; Hou, Z. Organometallics 2011, 30, 2908−
2919.
REFERENCES
■
(1) Facts & Figures of the Chemical Industry.Chem. Eng. News 2010,
88, 33−67.
(2) Nexant Process Evaluation Research Planning (PERP) Report-
Ethylene 08/09-5, 2009.
(3) Nexant Process Evaluation Research Planning (PERP) Report-Alpha
Olefins 06/07-5, 2008.
(4) Vogt, D. Oligomerization of Ethylene to Higher Linear α-Olefins.
In Applied Homogeneous Catalysis with Organometallic Compounds;
Cornils, B., Herrmann, W. A., Eds.; VCH: New York, 1996; Vol. 1, pp
245−258.
(5) Dixon, J. T.; Green, M. J.; Hess, F. M.; Morgan, D. H. J. Organomet.
Chem. 2004, 689, 3641−3668.
(6) McGuinness, D. S. Chem. Rev. 2011, 111, 2321−2341.
(7) Agapie, T. Coord. Chem. Rev. 2011, 255, 861−880.
(8) Wass, D. F. Dalton Trans. 2007, 816−819.
(9) Manyik, R. M.; Walker, W. E.; Wilson, T. P. U.S. Patent 3 300 458,
1967.
(35) Nelson, J. E.; Bercaw, J. E.; Labinger, J. A. Organometallics 1989, 8,
2484−2486.
(36) Terao, J.; Begum, S. A.; Oda, A.; Kambe, N. Synlett 2005, 1783−
1786.
(37) Kaisare, A. A.; Owens, S. B., Jr.; Valente, E. J.; Gray, G. M. J.
Organomet. Chem. 2010, 695, 1472−1479.
(38) Maura, R.; Steele, J.; Vendier, L.; Arquier, D.; Bastin, S.;
Urrutigoïty, M.; Kalck, P.; Igau, A. J. Organomet. Chem. 2011, 696, 897−
904.
(39) Nozaki, K.; Komaki, H.; Kawashima, Y.; Hiyama, T.; Matsubara,
T. J. Am. Chem. Soc. 2001, 123, 534−544.
(40) Suzuki, Y.; Kinoshita, S.; Shibahara, A.; Ishii, S.; Kawamura, K.;
Inoue, Y.; Fujita, T. Organometallics 2010, 29, 2394−2396.
(41) Wohl, A.; Muller, W.; Peulecke, N.; Muller, B. H.; Peitz, S.; Heller,
̈
̈
̈
D.; Rosenthal, U. J. Mol. Catal. A 2009, 297, 1−8.
(42) Wohl, A.; Muller, W.; Peitz, S.; Peulecke, N.; Aluri, B. R.; Muller,
B. H.; Heller, D.; Rosenthal, U.; Al-Hazmi, M. H.; Mosa, F. M. Chem.
Eur. J. 2010, 16, 7833−7842.
̈
̈
̈
(10) Manyik, R. M.; Walker, W. E.; Wilson, T. P. J. Catal. 1977, 47,
197−209.
(11) Briggs, J. R. Chem. Commun. 1989, 674−675.
G
dx.doi.org/10.1021/om300492r | Organometallics XXXX, XXX, XXX−XXX